CN109993743B - Vascular image processing method, device, equipment and storage medium - Google Patents
Vascular image processing method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN109993743B CN109993743B CN201910280712.6A CN201910280712A CN109993743B CN 109993743 B CN109993743 B CN 109993743B CN 201910280712 A CN201910280712 A CN 201910280712A CN 109993743 B CN109993743 B CN 109993743B
- Authority
- CN
- China
- Prior art keywords
- image
- initial
- target
- point
- edge curve
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000003672 processing method Methods 0.000 title claims abstract description 28
- 230000002792 vascular Effects 0.000 title claims description 19
- 210000004204 blood vessel Anatomy 0.000 claims abstract description 192
- 238000009499 grossing Methods 0.000 claims abstract description 39
- 238000000034 method Methods 0.000 claims abstract description 19
- 238000013519 translation Methods 0.000 claims description 33
- 238000012545 processing Methods 0.000 claims description 24
- 238000003384 imaging method Methods 0.000 claims description 16
- 230000008859 change Effects 0.000 claims description 10
- 238000012216 screening Methods 0.000 claims description 10
- 238000004590 computer program Methods 0.000 claims description 3
- 238000003745 diagnosis Methods 0.000 abstract description 6
- 230000029058 respiratory gaseous exchange Effects 0.000 abstract description 4
- 210000001715 carotid artery Anatomy 0.000 description 9
- 238000010586 diagram Methods 0.000 description 9
- 230000000241 respiratory effect Effects 0.000 description 9
- 238000004422 calculation algorithm Methods 0.000 description 8
- 238000001514 detection method Methods 0.000 description 7
- 238000002059 diagnostic imaging Methods 0.000 description 6
- 238000002604 ultrasonography Methods 0.000 description 6
- 230000007246 mechanism Effects 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 230000002902 bimodal effect Effects 0.000 description 3
- 238000012937 correction Methods 0.000 description 3
- 238000013500 data storage Methods 0.000 description 3
- 238000003708 edge detection Methods 0.000 description 3
- 230000001154 acute effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 239000008280 blood Substances 0.000 description 1
- 210000004369 blood Anatomy 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000001121 heart beat frequency Effects 0.000 description 1
- 230000003902 lesion Effects 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 230000008707 rearrangement Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 208000024891 symptom Diseases 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10132—Ultrasound image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30101—Blood vessel; Artery; Vein; Vascular
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Ultra Sonic Daignosis Equipment (AREA)
Abstract
The embodiment of the invention discloses a blood vessel image processing method, a device, equipment and a storage medium, wherein the method comprises the following steps: acquiring a vessel longitudinal cutting image, wherein the vessel longitudinal cutting image comprises a first initial image containing the upper wall of a vessel and a second initial image containing the lower wall of the vessel; determining turning points in an edge curve of the upper wall of the blood vessel in the first initial image, and performing smoothing treatment on the edge curve according to the position relation among the turning points to obtain a first reference image; rotating the first reference image to enable a target starting point and a target ending point of an edge curve of the first reference image to be respectively and correspondingly overlapped with an initial starting point and an initial ending point of the edge curve in the first initial image; and obtaining a target image according to the rotated first reference image and the second initial image. According to the technical scheme provided by the embodiment of the invention, under the condition that the integral posture of the original blood vessel is kept, the shaking noise caused by breathing and heartbeat is removed, and the reliability of the diagnosis result obtained based on the blood vessel image is improved.
Description
Technical Field
The embodiment of the invention relates to the technical field of image processing, in particular to a blood vessel image processing method, a blood vessel image processing device, blood vessel image processing equipment and a storage medium.
Background
Carotid ultrasound is an auxiliary examination for detecting whether an arterial vessel is abnormal or not, and is one of effective means for diagnosing and evaluating carotid wall lesions. However, when the carotid artery of the object to be detected is scanned, the flow of blood changes along with the change of the heartbeat frequency, so that the detected blood vessel image becomes thicker or thinner correspondingly, and the upper wall of the blood vessel of the longitudinal cutting graph of the carotid artery blood vessel is dithered, which is not beneficial to the observation and effective diagnosis of the symptoms by medical staff.
In the prior art, a decorrelation algorithm or a convolutional neural network and the like are generally adopted to carry out smoothing processing on an acquired blood vessel ultrasonic image, so that jitter noise caused by respiratory heartbeat is reduced. However, the acquired blood vessel ultrasonic images are processed by adopting the modes, so that the blood vessel trend of the original image is changed, and the reliability of the diagnosis result is necessarily affected by adopting the images with the changed blood vessel trend to diagnose and treat diseases.
Disclosure of Invention
The invention provides a blood vessel image processing method, a device, equipment and a storage medium, which are used for removing shaking noise caused by respiratory heartbeat and keeping the whole posture of an original blood vessel.
In a first aspect, an embodiment of the present invention provides a blood vessel image processing method, including:
acquiring a vessel longitudinal map, the vessel longitudinal map comprising a first initial image comprising an upper vessel wall and a second initial image comprising a lower vessel wall;
determining at least two turning points in an edge curve of the upper wall of the blood vessel in the first initial image, and performing smoothing treatment on the edge curve according to the position relationship between the at least two turning points to obtain a first reference image corresponding to the first initial image, wherein the curve sections on two sides of the turning points have opposite directions;
rotating the first reference image so that a target starting point of an edge curve of the first reference image corresponds to an initial starting point of an edge curve in the first initial image, and a target ending point of the edge curve of the first reference image corresponds to an initial ending point of the edge curve in the first initial image;
and obtaining a target image according to the rotated first reference image and the second initial image.
In a second aspect, an embodiment of the present invention further provides a blood vessel image processing apparatus, including:
an image acquisition module for acquiring a vessel slit map comprising a first initial image comprising an upper vessel wall and a second initial image comprising a lower vessel wall;
The smoothing processing module is used for determining at least two turning points in the edge curve of the upper wall of the blood vessel in the first initial image, and carrying out smoothing processing on the edge curve according to the position relation between the at least two turning points to obtain a first reference image corresponding to the first initial image, wherein the trend of curve sections at two sides of the turning points is opposite;
the image rotation module is used for rotating the first reference image so that a target starting point of an edge curve of the first reference image is correspondingly overlapped with an initial starting point of the edge curve in the first initial image, and a target ending point of the edge curve of the first reference image is correspondingly overlapped with an initial ending point of the edge curve in the first initial image;
and the target image obtaining module is used for obtaining a target image according to the rotated first reference image and the second initial image.
In a third aspect, an embodiment of the present invention further provides an electronic device, including:
one or more processors;
a memory for storing one or more programs,
the one or more programs, when executed by the one or more processors, cause the one or more processors to implement a vascular image processing method as provided by the embodiments of the first aspect.
In a fourth aspect, embodiments of the present invention also provide a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements a blood vessel image processing method as provided by the embodiments of the first aspect.
The embodiment of the invention acquires a blood vessel longitudinal cutting image comprising a first initial image of the upper wall of a blood vessel and a second initial image of the lower wall of the blood vessel; determining at least two turning points in an edge curve of the upper wall of the blood vessel in a first initial image, and performing smoothing treatment on the edge curve according to the position relation between the at least two turning points to obtain a first reference image corresponding to the first initial image; wherein the curve sections at two sides of the turning point have opposite directions; rotating the first reference image so that a target starting point of an edge curve of the first reference image corresponds to an initial starting point of an edge curve in the first initial image, and a target ending point of the edge curve of the first reference image corresponds to an initial ending point of the edge curve of the first initial image; and obtaining a target image according to the rotated first reference image and second reference image. According to the technical scheme, the sawtooth-shaped turning parts of the edge curve of the upper wall of the blood vessel caused by motion artifacts generated by respiratory heartbeat and the like are subjected to smoothing, and the rotation correction is carried out on the image obtained after the smoothing, so that the shaking noise caused by respiratory heartbeat is removed under the condition that the integral posture of the original blood vessel is kept, and the reliability of the diagnosis result obtained based on the blood vessel image is improved.
Drawings
FIG. 1A is a flow chart of a blood vessel image processing method according to a first embodiment of the present invention;
FIG. 1B is a longitudinal view of a blood vessel corresponding to an ultrasound image of a carotid artery in accordance with the first embodiment of the invention;
FIG. 1C is a first initial image in accordance with a first embodiment of the present invention;
FIG. 2A is a flow chart of a blood vessel image processing method in a second embodiment of the invention;
FIG. 2B is a first initial image in a second embodiment of the invention;
FIG. 3 is a flowchart of a blood vessel image processing method in accordance with a third embodiment of the present invention;
FIG. 4A is a flowchart of a vascular image processing method in accordance with a fourth embodiment of the present invention;
FIG. 4B is a second reference image in a fourth embodiment of the invention;
FIG. 5A is a flowchart of a vascular image processing method in a fifth embodiment of the present invention;
FIG. 5B is a longitudinal view of the carotid artery in embodiment five of the invention;
FIG. 5C is an upper image of a blood vessel in a fifth embodiment of the invention;
FIG. 5D is a schematic diagram of an edge curve in a fifth embodiment of the present invention;
FIG. 5E is a diagram of a region of interest image determination process in a fifth embodiment of the present invention;
fig. 6 is a block diagram of a blood vessel image processing apparatus in a sixth embodiment of the present invention;
fig. 7 is a schematic structural diagram of an electronic device in a seventh embodiment of the present invention.
Detailed Description
The invention is described in further detail below with reference to the drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting thereof. It should be further noted that, for convenience of description, only some, but not all of the structures related to the present invention are shown in the drawings.
Example 1
Fig. 1A is a flowchart of a blood vessel image processing method according to a first embodiment of the present invention. The embodiment of the invention is suitable for removing the dithering noise caused by breathing and heartbeat in the acquired blood vessel ultrasonic image. The method is performed by a vascular image processing device, which is implemented in software and/or hardware, and is specifically configured in an electronic device with image processing capability, where the electronic device may be a stand-alone computing device, such as a personal computer or a PC, or may be a data processing device included in a medical imaging system, where the medical imaging device may be an ultrasound diagnostic device.
A blood vessel image processing method as shown in fig. 1A, comprising:
s110, acquiring a blood vessel longitudinal cutting image, wherein the blood vessel longitudinal cutting image comprises a first initial image containing the upper wall of a blood vessel and a second initial image containing the lower wall of the blood vessel.
The blood vessel longitudinal cutting image is a longitudinal cutting image obtained by projecting the obtained three-dimensional blood vessel image along a plane in which the blood vessel shaking direction is located. Wherein the three-dimensional vessel image is generated based on a series of vessel two-dimensional scans. Wherein the blood vessel may be an arterial vessel, typically a carotid artery, affected by a respiratory heartbeat, which may cause vascular jitter; correspondingly, the vessel slit map is typically an image obtained by projecting an ultrasound image of the carotid artery along a plane in which the vessel is dithered. Wherein, the upper wall of the blood vessel can be understood as a blood vessel wall with relatively obvious blood vessel jitter in a blood vessel longitudinal cutting graph; the lower wall of a vessel is understood to be a vessel wall in which the vessel jitter is relatively insignificant in the vessel longitudinal map. Illustratively, fig. 1B shows a longitudinal view of a blood vessel corresponding to an ultrasound image of the carotid artery.
Alternatively, the obtaining of the vessel longitudinal map may be directly obtained from a local electronic device, another storage device associated with the electronic device, or a cloud end, and the vessel longitudinal map includes a first initial image including an upper wall of the vessel and a second initial image including a lower wall of the vessel.
Or alternatively, acquiring a vascular longitudinal map may be acquiring an original vascular longitudinal map; detecting a blood vessel region in the original blood vessel longitudinal cutting graph according to a blood vessel detection algorithm; dividing the original vascular longitudinal cutting graph according to the detected vascular region to obtain a first initial image and a second initial image, wherein the jitter condition of the vascular wall in the first initial image is obvious compared with that of the vascular wall in the second initial image. The blood vessel detection algorithm can be a threshold detection method, a maximum inter-class variance method, pattern recognition and the like. Illustratively, according to a blood vessel detection algorithm, detecting a blood vessel region in an original blood vessel longitudinal cutting image, which may be counting image gray values in the original blood vessel longitudinal cutting image, calculating an image binarization threshold according to a bimodal method, and performing edge detection and connected domain detection on the image according to the binarization threshold, and obtaining the blood vessel region after excluding irrelevant connected domains.
Alternatively, the first initial image and the second initial image may be divided by manually identifying, where an image with relatively obvious vascular wall shake is determined as the first initial image, and an image with relatively insignificant vascular wall shake is determined as the second initial image. Illustratively, fig. 1B shows an upper vessel wall a in which the vessel wall shake is relatively noticeable and a lower vessel wall B in which the vessel wall shake is relatively insignificant in the vessel longitudinal cut map.
Or alternatively, the division of the first initial image and the second initial image may be a first fluctuation parameter and a second fluctuation parameter that identify each pixel point included in the blood vessel edge curve of each initial image; if the first fluctuation parameter is larger than the second fluctuation parameter, determining an initial image corresponding to the first fluctuation parameter as a first initial image, and determining an initial image corresponding to the second fluctuation parameter as a second initial image; if the first fluctuation parameter is smaller than the second fluctuation parameter, determining that the initial image corresponding to the first fluctuation parameter is the second initial image, and determining that the initial image corresponding to the second fluctuation parameter is the first initial image.
S120, determining at least two turning points in an edge curve of the upper wall of the blood vessel in the first initial image, and carrying out smoothing treatment on the edge curve according to the position relation between the at least two turning points to obtain a first reference image corresponding to the first initial image.
Wherein, the curve sections at two sides of the turning point have opposite directions. See schematic representation of the upper wall of the blood vessel in the first initial image shown in fig. 1C. Wherein the turning points are T1, T2, …, T7 and T8. Wherein the curve trend of the T2T3 curve segment is opposite to that of the T3T4 curve segment.
It can be understood that, when detecting the edge curve of the blood vessel wall in the first initial image, due to the accuracy of the detection algorithm, a certain deviation may exist between the detected edge curve and the actual edge curve, and after detecting the edge curve of the blood vessel wall in the first initial image, before determining at least two turning points in the edge curve of the blood vessel upper wall in the first initial image, acquiring one pixel point as a current pixel point for each pixel point in the edge curve; each pixel point in the preset adjacent area containing the current pixel point in the radial direction is determined to be the reference pixel point, wherein the pixel point with the largest pixel difference value change with the adjacent pixel point is determined; and replacing each pixel point in the edge curve with a reference pixel point corresponding to each pixel point so as to update the edge curve. Wherein the radial direction may be understood as the y-axis direction in the first initial image. The preset neighborhood may be a pixel area corresponding to a set number of pixels, with the current pixel as a center, and taken up and down along the y-axis direction. The number of settings may be set by a skilled person based on an empirical value, and may be 10, for example.
It can be understood that, due to the existence of jitter, at least two turning points exist in the edge curve, and each curve segment in the edge curve is evaluated according to the positional relationship between the at least two turning points, so that the edge curve is smoothed, and a first reference image corresponding to a first initial image containing the edge curve is obtained.
S130, rotating the first reference image so that a target starting point of an edge curve of the first reference image corresponds to an initial starting point of an edge curve in the first initial image, and a target ending point of the edge curve of the first reference image corresponds to an initial ending point of the edge curve in the first initial image.
After the edge curve in the first initial image is subjected to smoothing treatment, the whole trend of the edge curve and the processed edge curve is changed, so that the whole posture of the blood vessel is changed. In order to ensure that the edge curve after the smoothing process and the edge curve before the smoothing process keep the overall posture of the blood vessel unchanged as much as possible, the first reference image with the posture changed is required to be rotated and adjusted, so that the target starting point of the edge curve of the first reference image is correspondingly overlapped with the initial starting point of the edge curve in the first initial image, and the target ending point of the edge curve of the first reference image is correspondingly overlapped with the initial ending point of the edge curve in the first initial image.
And S140, obtaining a target image according to the rotated first reference image and the second initial image.
Since the edge curve in the rotated first reference image is the same as the integral posture of the upper wall of the original blood vessel, and the dithering noise caused by the breathing and heartbeat is eliminated, the integral posture of the blood vessel combined by the upper wall and the lower wall of the blood vessel in the target image obtained according to the rotated first reference image and the second reference image is consistent with the blood vessel posture in the acquired blood vessel longitudinal cutting graph, and the dithering noise is eliminated.
The embodiment of the invention acquires a blood vessel longitudinal cutting image comprising a first initial image of the upper wall of a blood vessel and a second initial image of the lower wall of the blood vessel; determining at least two turning points in an edge curve of the upper wall of the blood vessel in a first initial image, and performing smoothing treatment on the edge curve according to the position relation between the at least two turning points to obtain a first reference image corresponding to the first initial image; wherein the curve sections at two sides of the turning point have opposite directions; rotating the first reference image so that a target starting point of an edge curve of the first reference image corresponds to an initial starting point of an edge curve in the first initial image, and a target ending point of the edge curve of the first reference image corresponds to an initial ending point of the edge curve of the first initial image; and obtaining a target image according to the rotated first reference image and second reference image. According to the technical scheme, the sawtooth-shaped turning parts of the edge curve of the upper wall of the blood vessel caused by motion artifacts generated by respiratory heartbeat and the like are subjected to smoothing, and the rotation correction is carried out on the image obtained after the smoothing, so that the shaking noise caused by respiratory heartbeat is removed under the condition that the integral posture of the original blood vessel is kept, and the reliability of the diagnosis result obtained based on the blood vessel image is improved.
Example two
Fig. 2A is a flowchart of a blood vessel image processing method in the second embodiment of the present invention. The embodiment of the invention is optimized and improved on the basis of the technical scheme of each embodiment.
Further, the operation of smoothing the edge curve according to the position relation between the at least two turning points is thinned into dividing the edge curve into at least three initial curve dividing sections by taking the at least two turning points as end points; screening at least two target curve segments from the initial curve segments according to the overall trend of the blood vessel; and taking two adjacent endpoints of the two adjacent target curve segments as a group of turning point pairs, and translating the target curve segments according to each group of turning point pairs so as to make the at least two target curve segments collinear to perfect a processing mode of smoothing the edge curve.
A blood vessel image processing method as shown in fig. 2A, comprising:
s210, acquiring a blood vessel longitudinal cutting image, wherein the blood vessel longitudinal cutting image comprises a first initial image containing the upper wall of a blood vessel and a second initial image containing the lower wall of the blood vessel.
S221, determining at least two turning points in an edge curve of the upper wall of the blood vessel in the first initial image.
Optionally, determining at least two turning points in the edge curve of the upper wall of the blood vessel in the first initial image may directly obtain location information of each turning point in the edge curve included in the first initial image from a storage device or cloud associated with the electronic device locally, where the location information may be a location coordinate of the turning point in the first initial image.
Or alternatively, determining at least two turning points in the edge curve of the vessel upper wall in the first initial image may be detecting the edge curve of the vessel wall in the first initial image according to a vessel detection algorithm; calculating the slope value of each adjacent pixel point (or points separated by N pixel values) in the edge curve, wherein the specific size of N is correspondingly adjusted according to the image size of a specific vascular longitudinal graph, and when the value of N is greater than 1, invalid turning points can be eliminated; and acquiring one of the pixel points as a current pixel point, and determining the current pixel point as a turning point if the slope value corresponding to the current pixel point is positive and negative. For example, the current pixel point is P i (x i ,y i ) Adjacent pixel points are P respectively i-1 (x i-1 ,y i-1 ) And P i+1 (x i+1 ,y i+1 ) If P i-1 P i Slope (y) i -y i-1 )/(x i -x i-1 ) And P i P i+1 Slope (y) i+1 -y i )/(x i+1 -x i ) If the positive and negative are opposite, determining the current pixel point P i Is the turning point.
S222, dividing the edge curve into at least three initial curve segments by taking the at least two turning points as end points.
The first initial image shown in fig. 2B is a schematic diagram, in which the curves including turning points T1, T2, …, T7, T8 are edge curves of the upper wall of the blood vessel. The edge curve is divided into at least three initial curve segments by each turning point.
S223, screening at least two target curve segments from the initial curve segments according to the overall trend of the blood vessel.
And screening initial curve segments which are consistent with the integral trend of the blood vessel from all the initial curve segments according to the integral trend of the blood vessel, and taking the initial curve segments as target curve segments. Referring to fig. 2B, from the overall trend of the upper wall of the blood vessel, it can be known that the overall posture of the blood vessel extends in the x-axis positive direction and the y-axis positive direction. And finally, screening out line segments AT1, T2T3, T4T5, T6T7 and T8B to obtain the target curve segmentation segment.
S224, two adjacent end points of the two adjacent target curve segments are taken as a group of turning point pairs, the target curve segments are translated according to each group of turning point pairs, and a first reference image corresponding to the first initial image is obtained, so that the at least two target curve segments are collinear.
Wherein, the curve sections at two sides of the turning point have opposite directions.
The target curve segments are made collinear by translating the target curve segments in the y-axis direction. In order to facilitate the determination of the translation distance and the translation direction, a translation reference point may be predetermined, and each target curve segment may be translated according to the reference point.
Illustratively, translating the target curve segment may be: determining adjacent turning points which are closer to each other along the extending direction of the blood vessel as turning point pairs, and determining one of other pixel points except for a segmented section formed by the turning point pairs as a translation reference point; for each turning point pair, one turning point pair is obtained as a current turning point pair, and the projection distance of a far turning point far away from a translation reference point in the current turning point pair in a projection straight line is determined; the projection straight line is a straight line formed by a near turning point which is close to the translation reference point in the current turning point and an adjacent turning point which is close to one side of the translation reference point; accumulating projection distances corresponding to the current turning point pair and other turning point pairs between the current turning time point pair and the translation reference point to obtain translation distances corresponding to the current turning point pair; and moving a to-be-translated line segment formed by a far turning point in the current turning point pair and an adjacent turning point on one side of the far turning point far away from the translation reference point along the radial direction close to the translation reference point by the translation distance until all the to-be-translated line segments in the edge curve are translated, and obtaining the first reference image. Wherein the radial direction corresponds to the y-axis direction in fig. 2B.
Specifically, referring to FIG. 2B, the determined turning point pairs are { T1, T2}, { T3, T4}, { T5, T6}, and { T7, T8}; selecting a pixel point O between T4 and T5 as a translation reference point; then, the far turning points in the turning point pair { T1, T2} and { T3, T4} are T1 and T3, and the near turning points are T2 and T4; the far turning points in the pair of turning points { T5, T6} and { T7, T8} are T6 and T8, and the near turning points are T5 and T7. The projection straight line corresponding to the far turning point T1 is T2T3, the line segment to be translated is AT1, the translation distance is (h1+h2), and the translation direction is the negative y-axis direction; the projection straight line corresponding to the far turning point T3 is T4T5, the line segment to be translated is T2T3, the translation distance is h2, and the translation direction is the negative y-axis direction; the projection straight line corresponding to the far turning point T6 is T4T5, the line segment to be translated is T6T7, the translation distance is h3, and the translation direction is the positive y-axis direction; the projection straight line corresponding to the far turning point T8 is T6T7, the line segment to be translated is T8B, the translation distance is (h3+h4), and the translation direction is the positive y-axis direction.
To facilitate the determination of the translation distance and translation direction, an initial start point or an initial end point of the edge curve may be determined as a translation reference point. Preferably, the initial starting point of the edge curve is determined as the translation reference point.
S230, rotating the first reference image so that a target starting point of an edge curve of the first reference image corresponds to an initial starting point of an edge curve in the first initial image, and a target ending point of the edge curve of the first reference image corresponds to an initial ending point of the edge curve in the first initial image.
Wherein, the initial starting point is the A point, the initial ending point is the B point, the target starting point is the A ', and the target ending point is the B'.
S240, obtaining a target image according to the rotated first reference image and the second initial image.
According to the embodiment of the invention, the edge curve is divided into at least three initial curve segments by thinning the operation of smoothing the edge curve according to the position relation between at least two turning points by taking the at least two turning points as endpoints; screening at least two target curve segments from the initial curve segments according to the overall trend of the blood vessel; and taking two adjacent endpoints of two adjacent target curve segments as a group of turning point pairs, and translating the target curve segments according to each group of turning point pairs so as to make at least two target curve segments collinear. By adopting the technical scheme, the extending trend of the upper wall of the blood vessel is reserved by translating the target curve segments contained in the edge curve, and the edge curve of the upper wall of the blood vessel is smoothed on the premise of not changing the integral posture of the upper wall of the blood vessel, so that the processing mode of smoothing the edge curve is further perfected.
Example III
Fig. 3 is a flowchart of a blood vessel image processing method in the third embodiment of the present invention. The embodiment of the invention is optimized and improved on the basis of the technical scheme of each embodiment.
Further, the operation of rotating the first reference image so that the target starting point of the edge curve of the first reference image corresponds to the initial starting point of the edge curve in the first initial image, and the corresponding superposition of the target ending point of the edge curve of the first reference image and the initial ending point of the edge curve in the first initial image is thinned to rotate the first reference image around the target starting point or the target ending point, so that the target starting point corresponds to the initial starting point, and the target ending point corresponds to the initial ending point, so as to perfect the rotation mode when the first reference image is rotated.
Further, before the operation of "rotating the first reference image", the rotation angle "when the first reference image is rotated is determined by adding" according to the positional relationship of the initial start point, the target start point, the initial end point, and the target end point, so as to perfect a determination mechanism of the rotation angle.
A blood vessel image processing method as shown in fig. 3, comprising:
s310, acquiring a blood vessel longitudinal cutting image, wherein the blood vessel longitudinal cutting image comprises a first initial image containing the upper wall of a blood vessel and a second initial image containing the lower wall of the blood vessel.
S320, determining at least two turning points in an edge curve of the upper wall of the blood vessel in the first initial image, and carrying out smoothing treatment on the edge curve according to the position relation between the at least two turning points to obtain a first reference image corresponding to the first initial image, wherein the trend of curve sections at two sides of the turning points is opposite.
S330, determining a rotation angle when the first reference image is rotated according to the position relation among the initial starting point, the target starting point, the initial ending point and the target ending point.
Specifically, according to the position coordinates of the initial starting point and the initial ending point, determining the coordinate value of an initial vector of a straight line where the initial starting point and the initial ending point are positioned; according to the position coordinates of the target starting point and the target ending point, determining the coordinate value of a target vector of a straight line where the target starting point and the target ending point are positioned; and determining an acute angle included between the initial vector and the target vector as a rotation angle according to the coordinate values of the initial vector and the coordinate values of the target vector.
S340, rotating the first reference image around the target starting point or the target ending point so that the target starting point and the initial starting point are correspondingly overlapped, and the target ending point and the initial ending point are correspondingly overlapped.
Optionally, translating the first reference image along the y-axis direction (see fig. 2B) such that the target starting point of the first reference image coincides with the initial starting point of the first initial image; taking a target starting point of the first reference image as a rotation reference point, and rotating the first reference image according to the determined rotation angle so as to enable a target ending point to coincide with an initial ending point;
or alternatively, the first reference image is translated in the y-axis direction (see fig. 2B) such that the target end point of the first reference image coincides with the initial end point of the first initial image; and rotating the first reference image according to the determined rotation angle by taking the target end point of the first reference image as a rotation reference point so as to enable the target start point to coincide with the initial start point.
Or alternatively, selecting any point in the edge curve in the first reference image as a rotation reference point; rotating the determined rotation angle around the rotation reference point to enable the edge curve in the rotated first reference image to be parallel to the initial vector corresponding to the edge curve in the first initial image; translating the rotated first reference image, so that the target starting point of the translated first reference image is correspondingly overlapped with the initial starting point of the first initial image, and the target ending point of the translated first reference image is correspondingly overlapped with the initial ending point of the first initial image.
In order to facilitate the execution of the rotation operation while reducing the operations performed when the target end point is made to coincide with the initial end point and the target start point is made to coincide with the initial start point, it is preferable to select the initial start point of the first initial image as the translation reference point at the time of translation; accordingly, the target start point of the first reference image is selected as the rotation reference point at the time of rotation.
S350, obtaining a target image according to the rotated first reference image and the second initial image.
According to the embodiment of the invention, through the operation of rotating the first reference image, the first reference image is thinned to rotate around the target starting point or the target ending point, so that the target starting point is correspondingly overlapped with the initial starting point, and the target ending point is correspondingly overlapped with the initial ending point, thereby perfecting the rotating mode when the first reference image is rotated; before the first reference image is rotated, the rotation angle of the first reference image is determined by additionally determining the position relation of the initial starting point, the target starting point, the initial ending point and the target ending point, so that a determination mechanism of the rotation angle is perfected, and further, a guarantee is provided for keeping the extending trend of the upper wall of the blood vessel.
Example IV
Fig. 4A is a flowchart of a blood vessel image processing method in the fourth embodiment of the present invention. The embodiment of the invention is optimized and improved on the basis of the technical scheme of each embodiment.
Further, the operation of obtaining a target image from the rotated first reference image and the second initial image is refined to "determining a second reference image from the rotated first reference image and the first initial image; and splicing the second reference image and the second initial image to obtain a target image so as to perfect a determination mechanism of the target image.
A blood vessel image processing method as shown in fig. 4A, comprising:
s410, acquiring a blood vessel longitudinal cutting image, wherein the blood vessel longitudinal cutting image comprises a first initial image containing the upper wall of a blood vessel and a second initial image containing the lower wall of the blood vessel.
S420, determining at least two turning points in an edge curve of the upper wall of the blood vessel in the first initial image, and carrying out smoothing treatment on the edge curve according to the position relation between the at least two turning points to obtain a first reference image corresponding to the first initial image, wherein the trend of curve sections at two sides of the turning points is opposite.
S430, rotating the first reference image so that the target starting point of the edge curve of the first reference image is correspondingly overlapped with the initial starting point of the edge curve in the first initial image, and the target ending point of the edge curve of the first reference image is correspondingly overlapped with the initial ending point of the edge curve in the first initial image.
S441, determining a second reference image from the rotated first reference image and the first initial image.
Optionally, determining a second reference image according to the rotated first reference image and the first initial image, which may be that a target imaging area of the second reference image is constructed by using the first initial image, and filling data of an overlapping area of the rotated first reference image and the first initial image into the target imaging area; filling each pixel point to be filled in an unfilled target imaging region outside the upper wall of a blood vessel according to the pixel point to be filled, and setting a pixel value of a radial pixel region; and filling each pixel point to be filled in the unfilled target imaging region on the inner side of the upper wall of the blood vessel according to the corresponding pixel value in the first initial image.
See the schematic diagram of the second reference image shown in fig. 4B. Specifically, the data of the overlapping region with the first initial image in the rotated first reference image is filled in the target imaging region 40. The target imaging region 41 outside the upper wall of the blood vessel is filled with pixel values of a radial pixel region set according to the pixel points to be filled. The radial pixel region is set to be a new pixel value calculated according to a setting algorithm according to the pixel value of the corresponding region in the first initial image; wherein, the setting algorithm can be set by the technicians according to the needs. The filling is preferably performed directly with pixel values of the corresponding areas in the first initial image. The target imaging region 42 and the target imaging region 43 inside the upper wall of the blood vessel are filled with pixel values of the corresponding regions in the first initial image.
Or alternatively, in order to simplify the calculation, the first initial image for clearly showing the edge curve of the upper wall of the blood vessel can be directly used as the second reference image; and filling the coordinates of each pixel point of the edge curve in the rotated first reference image into the second reference image in sequence according to the rotated position coordinates. It will be appreciated that in order to ensure complete filling of the pixel values of the pixels of the smoothed and rotated edge curve, it is preferable to replace the corresponding pixels in the second reference image with the smoothed and rotated edge curve pixels when overlapping the smoothed and rotated edge curve pixels with the pixels in the second reference image.
S442, splicing the second reference image and the second initial image to obtain a target image.
According to the embodiment of the invention, the operation of obtaining the target image according to the rotated first reference image and second initial image is refined into the determination of the second reference image according to the rotated first reference image and first initial image; and splicing the second reference image and the second initial image to obtain a target image, perfecting a determination mechanism of the target image, and further providing a guarantee for displaying the complete blood vessel image.
Example five
Fig. 5A is a flowchart of a blood vessel image processing method in a fifth embodiment of the present invention. The embodiment of the invention provides a preferred implementation mode based on the technical scheme of each embodiment.
A blood vessel image processing method as shown in fig. 5A, comprising:
s501, acquiring an ultrasonic three-dimensional image of the carotid artery, and determining a longitudinal cutting image of the carotid artery according to the blood vessel shaking direction.
Wherein the carotid longitudinal view is seen in fig. 5B.
S502, detecting the position of a blood vessel in the carotid longitudinal map, and dividing the carotid longitudinal map image into an upper blood vessel image and a lower blood vessel image according to the position of the blood vessel.
Specifically, counting the image gray value in the carotid longitudinal graph, and calculating the binarization threshold of the carotid longitudinal graph according to a bimodal method; performing edge detection on the image according to the binarization threshold, detecting through the connected domain, removing irrelevant connected domains, and determining the position of a blood vessel according to the position of the connected domain; dividing the carotid longitudinal map into two partial images each comprising a vessel wall according to vessel position; recognizing the shaking condition of the blood vessel wall in the two partial images, and determining the partial image with obvious shaking of the blood vessel wall as an upper image of the blood vessel; the image in which the vessel wall shake is not significant is a lower image of the vessel. Wherein the upper image of the blood vessel is seen in fig. 5C.
S503, detecting an edge curve in the upper image of the blood vessel.
Specifically, the gray value of an upper image of a blood vessel is counted, a binarization threshold value of the upper image of the blood vessel is calculated according to a bimodal method, edge detection is carried out on the upper image of the blood vessel according to the binarization threshold value, and a first non-zero pixel point in the direction from the inner side of the blood vessel wall to the outer side of the blood vessel wall in the detected image is identified as an initial pixel point; and combining all the initial pixel points to obtain an edge curve.
S504, according to gray scale change conditions of adjacent pixel points in the setting neighborhood of each pixel point in the initial edge curve, determining a point with the largest change rate to replace the current pixel point so as to update the edge curve.
Specifically, 10 pixel points above and below the initial pixel point are obtained to obtain a reference pixel point corresponding to the initial pixel point; and according to the gray value change between every two adjacent reference pixel points, determining the reference pixel point with the maximum gray value change rate to replace the corresponding initial pixel point in the edge curve so as to update the edge curve.
S505, determining turning points in the edge curve according to the position change of each pixel point in the updated edge curve. Wherein the change trend of curve sections at two sides of the turning point is different.
See the edge curve schematic shown in fig. 5D.
S506, taking the starting point of the edge curve as a reference point, and grouping the turning points in pairs along the extending direction of the blood vessel to obtain turning point pairs, wherein the turning point pairs comprise a starting turning point and a stopping turning point.
S507, calculating translation distances of the initial turning point and the termination turning point of the turning point pair in the radial direction of the blood vessel according to the turning point pairs; and translating the ending turning point and the pixel points after the ending turning point along the direction that the blood vessel is radially close to the starting turning point by the translation distance to obtain a blood vessel reference image corresponding to the image on the upper part of the blood vessel.
S508, determining the rotation angle of the upper image of the blood vessel according to the positions of the starting point and the ending point of the edge curve before and after the translation.
Specifically, according to an initial vector corresponding to a start point and an end point of an edge curve of the translation line and a target vector corresponding to the start point and the end point of the translated edge curve, determining an acute angle included angle between the initial vector and the target vector as a rotation angle.
See fig. 5D, where the initial vector is AB, the target vector is AB', and the rotation angle is α.
Specifically, the rotation angle is determined according to the following formula:
wherein, (x) 1 ,y 1 ) Is the vector coordinates of vector AB' (x) 2 ,y 2 ) Is the vector coordinates of vector AB.
S509, rotating the blood vessel reference graph according to the determined rotation angle, so that the starting point and the ending point of the edge curve in the blood vessel reference graph are respectively overlapped with the starting point and the ending point in the upper image of the blood vessel.
S510, determining an area-of-interest image according to the rotated blood vessel reference image and the blood vessel upper image, and splicing the area-of-interest image with the blood vessel lower image to obtain a blood vessel smoothed image.
See the region of interest image determination process diagram shown in fig. 5E.
Rotating the blood vessel reference map around the starting point A to obtain a temporary image NHTF; determining the minimum circumscribed rectangle LMQG of the temporary image; determining a reference image KPGF according to coordinates of two points NF positioned on the outer side of a blood vessel wall in the temporary image NHTF; overlapping the vessel upper image with the reference image so that the edge curve in the reference image coincides with the initial vector in the vessel upper image, and determining a region of interest image A1A2C 1; and splicing the region of interest image with the lower blood vessel image to obtain a smoothed blood vessel image, and cutting the part of the lower blood vessel image, which exceeds the region of interest image.
Example six
Fig. 6 is a block diagram of a blood vessel image processing apparatus in a sixth embodiment of the present invention. The embodiment of the invention is suitable for removing the dithering noise caused by breathing and heartbeat in the acquired blood vessel ultrasonic image. The device is realized by software and/or hardware, and is specifically configured in an electronic device with image processing capability, wherein the electronic device can be an independent computing device, such as a personal computer or a PC (personal computer) or the like, and can also be a data processing device contained in a medical imaging system, and the medical imaging device can be an ultrasonic diagnostic device.
A blood vessel image processing apparatus as shown in fig. 6, comprising: an image acquisition module 610, a smoothing module 620, an image rotation module 630, and a target image acquisition module 640.
Wherein the image acquisition module 610 is configured to acquire a blood vessel longitudinal graph, where the blood vessel longitudinal graph includes a first initial image including an upper wall of a blood vessel and a second initial image including a lower wall of the blood vessel;
a smoothing module 620, configured to determine at least two turning points in an edge curve of an upper wall of a blood vessel in the first initial image, and perform smoothing on the edge curve according to a positional relationship between the at least two turning points to obtain a first reference image corresponding to the first initial image, where curve sections on two sides of the turning points have opposite directions;
an image rotation module 630, configured to rotate the first reference image so that a target start point of an edge curve of the first reference image corresponds to an initial start point of an edge curve in the first initial image, and a target end point of an edge curve of the first reference image corresponds to an initial end point of an edge curve in the first initial image;
And a target image obtaining module 640, configured to obtain a target image according to the rotated first reference image and the second initial image.
According to the embodiment of the invention, a first initial image comprising the upper wall of a blood vessel and a second initial image comprising the lower wall of the blood vessel are acquired through an image acquisition module; determining at least two turning points in an edge curve of the upper wall of the blood vessel in a first initial image through a smoothing processing module, and carrying out smoothing processing on the edge curve according to the position relation between the at least two turning points to obtain a first reference image corresponding to the first initial image; wherein the curve sections at two sides of the turning point have opposite directions; rotating the first reference image through the image rotation module so that a target starting point of an edge curve of the first reference image corresponds to an initial starting point of the edge curve in the first initial image, and a target ending point of the edge curve of the first reference image corresponds to an initial ending point of the edge curve of the first initial image; and obtaining a target image according to the rotated first reference image and the second initial image by a target image obtaining module. According to the technical scheme, the sawtooth-shaped turning parts of the edge curve of the upper wall of the blood vessel caused by motion artifacts generated by respiratory heartbeat and the like are subjected to smoothing, and the rotation correction is carried out on the image obtained after the smoothing, so that the shaking noise caused by respiratory heartbeat is removed under the condition that the integral posture of the original blood vessel is kept, and the reliability of the diagnosis result obtained based on the blood vessel image is improved.
Further, the smoothing module 620, when performing a smoothing operation on the edge curve according to the positional relationship between the at least two turning points, specifically includes:
the dividing unit is used for dividing the edge curve into at least three initial curve dividing sections by taking the at least two turning points as end points;
the screening unit is used for screening at least two target curve segments from the initial curve segments according to the overall trend of the blood vessel;
and the translation unit is used for translating the two adjacent endpoints of the two adjacent target curve segments into a group of turning point pairs according to each group of turning point pairs so as to enable the at least two target curve segments to be collinear.
Further, the image rotation module 630 includes:
and the image rotating unit is used for rotating the first reference image around the target starting point or the target ending point so as to enable the target starting point to be correspondingly overlapped with the initial starting point and enable the target ending point to be correspondingly overlapped with the initial ending point.
Further, the image rotation module 630 further includes:
and the rotation angle determining unit is used for determining the rotation angle when the first reference image is rotated according to the position relation among the initial starting point, the target starting point, the initial ending point and the target ending point before the first reference image is rotated.
Further, the device further comprises an edge curve determining module, which is specifically configured to:
detecting an edge curve of a vessel upper wall in the first initial image before determining at least two turning points in the edge curve of the vessel upper wall in the first initial image;
for each pixel point in the edge curve, acquiring one pixel point as a current pixel point;
each pixel point in the preset adjacent area containing the current pixel point in the radial direction is determined to be the reference pixel point, wherein the pixel point with the largest pixel difference value change with the adjacent pixel point is determined;
and replacing each pixel point in the edge curve with a reference pixel point corresponding to each pixel point so as to update the edge curve.
Further, the target image obtaining module 640 includes:
an image determining unit configured to determine a second reference image from the rotated first reference image and the first initial image;
and the image stitching unit is used for stitching the second reference image and the second initial image to obtain a target image.
Further, the image determining unit is specifically configured to:
constructing a target imaging area of a second reference image by the first initial image, and filling data of an overlapping area of the rotated first reference image and the first initial image into the target imaging area;
Filling each pixel point to be filled in an unfilled target imaging region outside the upper wall of a blood vessel according to the pixel point to be filled, and setting a pixel value of a radial pixel region;
and filling each pixel point to be filled in the unfilled target imaging region on the inner side of the upper wall of the blood vessel according to the corresponding pixel value in the first initial image.
The blood vessel image determining device can execute the blood vessel image determining method provided by any embodiment of the invention, and has the corresponding functional modules and beneficial effects of executing the blood vessel image determining method.
Example seven
Fig. 7 is a schematic structural diagram of an electronic device according to a seventh embodiment of the present invention, where the electronic device includes: processor 710 and storage 720.
One or more processors 710;
a storage 720 for storing one or more programs.
The electronic device further includes:
an input device 730 for receiving a vascular longitudinal map;
and an output device 740 for displaying the image.
The electronic device may be a stand-alone computing device, such as a personal computer or a PC, or may be a data processing device included in a medical imaging system, where the medical imaging device may be an ultrasound diagnostic device.
In fig. 7, for example, a processor 710 is shown, an input device 730 in the electronic device may be connected to an output device 740, the processor 710, and a memory device 720 by a bus or other means, and the processor 710 and the memory device 720 are also connected by a bus or other means, for example, in fig. 7.
In this embodiment, the processor 710 in the electronic device may directly acquire the blood vessel longitudinal map including the first initial image including the upper wall of the blood vessel and the second initial image including the lower wall of the blood vessel from the storage device 720, or acquire the blood vessel longitudinal map through the input device 730; at least two turning points in an edge curve of the upper wall of the blood vessel in a first initial image can be determined, and smoothing is carried out on the edge curve according to the position relation between the at least two turning points, so that a first reference image corresponding to the first initial image is obtained; the first reference image can be rotated so that the target starting point of the edge curve of the first reference image corresponds to the initial starting point of the edge curve in the first initial image, and the target ending point of the edge curve of the first reference image corresponds to the initial ending point of the edge curve in the first initial image; the target image can also be obtained according to the rotated first reference image and the second initial image.
The storage device 720 in the electronic device is used as a computer readable storage medium, and may be used to store one or more programs, such as a software program, a computer executable program, and a module, for example, program instructions/modules corresponding to the blockchain-based data storage method in the blockchain-based data storage method according to the embodiment of the present invention (for example, the image acquisition module 610, the smoothing module 620, the image rotation module 630, and the target image obtaining module 640 shown in fig. 6). The processor 710 executes various functional applications of the electronic device and data processing, that is, implements the blood vessel image processing method in the above-described method embodiment, by running the software programs, instructions, and modules stored in the storage 720.
The storage 720 may include a storage program area and a storage data area, wherein the storage program area may store an operating system, at least one application program required for a function; the stored data area may store data and the like (such as a vessel slitting map, a first initial image, an edge curve, a first reference image, a turning point, a second initial image, a target image, and the like in the above-described embodiment). In addition, the storage 720 may include high-speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid-state storage device. In some examples, storage 720 may further include memory located remotely from processor 710, which may be connected to the server via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
Example eight
An eighth embodiment of the present invention also provides a computer readable storage medium having stored thereon a computer program which, when executed by a blockchain-based data storage device, implements a blood vessel image processing method provided by the implementation of the present invention, including: acquiring a vessel longitudinal map, the vessel longitudinal map comprising a first initial image comprising an upper vessel wall and a second initial image comprising a lower vessel wall; determining at least two turning points in an edge curve of the upper wall of the blood vessel in the first initial image, and performing smoothing treatment on the edge curve according to the position relationship between the at least two turning points to obtain a first reference image corresponding to the first initial image, wherein the curve sections on two sides of the turning points have opposite directions; rotating the first reference image so that a target starting point of an edge curve of the first reference image corresponds to an initial starting point of an edge curve in the first initial image, and a target ending point of the edge curve of the first reference image corresponds to an initial ending point of the edge curve in the first initial image; and obtaining a target image according to the rotated first reference image and the second initial image.
Note that the above is only a preferred embodiment of the present invention and the technical principle applied. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, while the invention has been described in connection with the above embodiments, the invention is not limited to the embodiments, but may be embodied in many other equivalent forms without departing from the spirit or scope of the invention, which is set forth in the following claims.
Claims (8)
1. A blood vessel image processing method, comprising:
acquiring a vessel longitudinal map, the vessel longitudinal map comprising a first initial image comprising an upper vessel wall and a second initial image comprising a lower vessel wall;
determining at least two turning points in an edge curve of the upper wall of the blood vessel in the first initial image, and performing smoothing treatment on the edge curve according to the position relationship between the at least two turning points to obtain a first reference image corresponding to the first initial image, wherein the curve sections on two sides of the turning points have opposite directions;
Rotating the first reference image so that a target starting point of an edge curve of the first reference image corresponds to an initial starting point of an edge curve in the first initial image, and a target ending point of the edge curve of the first reference image corresponds to an initial ending point of the edge curve in the first initial image;
obtaining a target image according to the rotated first reference image and the second initial image;
smoothing the edge curve according to the position relation between the at least two turning points, including:
dividing the edge curve into at least three initial curve segments by taking the at least two turning points as end points;
screening at least two target curve segments from the initial curve segments according to the overall trend of the blood vessel;
two adjacent end points of two adjacent target curve segments are taken as a group of turning point pairs, and the target curve segments are shifted according to each group of turning point pairs so as to enable the at least two target curve segments to be collinear;
rotating the first reference image such that a target start point of an edge curve of the first reference image corresponds to a start point of an edge curve in the first initial image, and such that a target end point of an edge curve of the first reference image corresponds to a start point of an edge curve in the first initial image, comprising:
And rotating the first reference image around the target starting point or the target ending point so as to enable the target starting point to be correspondingly overlapped with the initial starting point and enable the target ending point to be correspondingly overlapped with the initial ending point.
2. The method of claim 1, further comprising, prior to rotating the first reference image:
and determining a rotation angle when the first reference image is rotated according to the position relation among the initial starting point, the target starting point, the initial ending point and the target ending point.
3. The method of claim 1, further comprising, prior to determining at least two turning points in an edge curve of a vessel upper wall in the first initial image:
detecting an edge curve of the upper wall of the blood vessel in the first initial image;
for each pixel point in the edge curve, acquiring one pixel point as a current pixel point;
each pixel point in the preset adjacent area containing the current pixel point in the radial direction is determined to be the reference pixel point, wherein the pixel point with the largest pixel difference value change with the adjacent pixel point is determined;
and replacing each pixel point in the edge curve with a reference pixel point corresponding to each pixel point so as to update the edge curve.
4. A method according to any one of claims 1-3, wherein obtaining a target image from the rotated first reference image and the second initial image comprises:
determining a second reference image according to the rotated first reference image and the first initial image;
and splicing the second reference image and the second initial image to obtain a target image.
5. The method of claim 4, wherein determining a second reference image from the rotated first reference image and the first initial image comprises:
constructing a target imaging area of a second reference image by the first initial image, and filling data of an overlapping area of the rotated first reference image and the first initial image into the target imaging area;
filling each pixel point to be filled in an unfilled target imaging region outside the upper wall of a blood vessel according to the pixel point to be filled, and setting a pixel value of a radial pixel region;
and filling each pixel point to be filled in the unfilled target imaging region on the inner side of the upper wall of the blood vessel according to the corresponding pixel value in the first initial image.
6. A blood vessel image processing apparatus, comprising:
An image acquisition module for acquiring a vessel slit map comprising a first initial image comprising an upper vessel wall and a second initial image comprising a lower vessel wall;
the smoothing processing module is used for determining at least two turning points in the edge curve of the upper wall of the blood vessel in the first initial image, and carrying out smoothing processing on the edge curve according to the position relation between the at least two turning points to obtain a first reference image corresponding to the first initial image, wherein the trend of curve sections at two sides of the turning points is opposite;
the image rotation module is used for rotating the first reference image so that a target starting point of an edge curve of the first reference image is correspondingly overlapped with an initial starting point of the edge curve in the first initial image, and a target ending point of the edge curve of the first reference image is correspondingly overlapped with an initial ending point of the edge curve in the first initial image;
the target image obtaining module is used for obtaining a target image according to the rotated first reference image and the second initial image;
the smoothing module, when executing the smoothing operation to the edge curve according to the position relation between the at least two turning points, specifically includes:
The dividing unit is used for dividing the edge curve into at least three initial curve dividing sections by taking the at least two turning points as end points;
the screening unit is used for screening at least two target curve segments from the initial curve segments according to the overall trend of the blood vessel;
the translation unit is used for translating the two adjacent endpoints of the two adjacent target curve segments into a group of turning point pairs according to each group of turning point pairs so as to enable the at least two target curve segments to be collinear;
an image rotation module comprising:
and the image rotating unit is used for rotating the first reference image around the target starting point or the target ending point so as to enable the target starting point to be correspondingly overlapped with the initial starting point and enable the target ending point to be correspondingly overlapped with the initial ending point.
7. An electronic device, the electronic device comprising:
one or more processors;
a memory for storing one or more programs,
when the one or more programs are executed by the one or more processors, the one or more processors are caused to implement a vascular image processing method as claimed in any one of claims 1-5.
8. A computer-readable storage medium, on which a computer program is stored, characterized in that the program, when being executed by a processor, implements a vascular image processing method as claimed in any one of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910280712.6A CN109993743B (en) | 2019-04-09 | 2019-04-09 | Vascular image processing method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910280712.6A CN109993743B (en) | 2019-04-09 | 2019-04-09 | Vascular image processing method, device, equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109993743A CN109993743A (en) | 2019-07-09 |
| CN109993743B true CN109993743B (en) | 2023-06-06 |
Family
ID=67132649
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910280712.6A Active CN109993743B (en) | 2019-04-09 | 2019-04-09 | Vascular image processing method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109993743B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112098055B (en) * | 2020-07-07 | 2021-03-23 | 青岛大学附属医院 | System and method for detecting blood transfusion pipeline shaking |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101833757A (en) * | 2009-03-11 | 2010-09-15 | 深圳迈瑞生物医疗电子股份有限公司 | Method and system for detection edge of blood vessel graphic tissue structure and blood vessel endangium |
| CN103824295A (en) * | 2014-03-03 | 2014-05-28 | 天津医科大学 | Segmentation method of adhesion vascular pulmonary nodules in lung CT (computed tomography) image |
| CN105976335A (en) * | 2016-05-09 | 2016-09-28 | 苏州佳世达电通有限公司 | Image processing method |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6845260B2 (en) * | 2001-07-18 | 2005-01-18 | Koninklijke Philips Electronics N.V. | Automatic vessel indentification for angiographic screening |
| US20100067753A1 (en) * | 2005-06-21 | 2010-03-18 | Koninklijke Philips Electronics, N.V. | Method and device for imaging a blood vessel |
| WO2014080961A1 (en) * | 2012-11-20 | 2014-05-30 | 株式会社 東芝 | Image processing device, image processing method and x-ray diagnosis device |
| CN108888284A (en) * | 2018-05-18 | 2018-11-27 | 沈阳东软医疗***有限公司 | Image adjusting method, device and equipment, storage medium |
-
2019
- 2019-04-09 CN CN201910280712.6A patent/CN109993743B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101833757A (en) * | 2009-03-11 | 2010-09-15 | 深圳迈瑞生物医疗电子股份有限公司 | Method and system for detection edge of blood vessel graphic tissue structure and blood vessel endangium |
| CN103824295A (en) * | 2014-03-03 | 2014-05-28 | 天津医科大学 | Segmentation method of adhesion vascular pulmonary nodules in lung CT (computed tomography) image |
| CN105976335A (en) * | 2016-05-09 | 2016-09-28 | 苏州佳世达电通有限公司 | Image processing method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109993743A (en) | 2019-07-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP2010088795A (en) | Medical image processor and medical image diagnosis device | |
| KR102560911B1 (en) | Image processing apparatus, image processing method, and storage medium | |
| US10542955B2 (en) | Method and apparatus for medical image registration | |
| JP6564018B2 (en) | Radiation image lung segmentation technology and bone attenuation technology | |
| JP5197140B2 (en) | X-ray fluoroscopic apparatus, moving image processing method, program, and storage medium | |
| JP5464799B2 (en) | Image processing apparatus, image processing method, and program | |
| CN107622501B (en) | Boundary detection method for medical image | |
| US10002424B2 (en) | Image processing system and method to reconstruct a three-dimensional (3D) anatomical surface | |
| CN110706791B (en) | Medical image processing method and device | |
| JP2007105264A (en) | Medical picture judgment apparatus, medical picture judgment method, and program thereof | |
| TW201545554A (en) | Image processing apparatus | |
| US9547906B2 (en) | System and method for data driven editing of rib unfolding | |
| JP2008043564A (en) | Domain definer | |
| CN109993743B (en) | Vascular image processing method, device, equipment and storage medium | |
| US20070053564A1 (en) | Image processing method and computer readable medium for image processing | |
| JP2002222410A (en) | Image diagnostic device | |
| US20020072665A1 (en) | Detection of ribcage boundary from digital chest image | |
| CN113344926B (en) | Method, device, server and storage medium for recognizing biliary-pancreatic ultrasonic image | |
| JP2007159933A (en) | Image display method, program and apparatus | |
| JP4738236B2 (en) | Image display device | |
| CN116363038A (en) | Ultrasonic image fusion method, device, computer equipment and storage medium | |
| JP2011110357A (en) | Image processing method, image processing apparatus and program | |
| US10413364B1 (en) | Internal organ localization of a subject for providing assistance during surgery | |
| CN114947941A (en) | Focus tracking method, device and storage medium | |
| US20220338819A1 (en) | Optimal scan pattern for medical imaging device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP01 | Change in the name or title of a patent holder | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 215000 5th floor, building a, 4th floor, building C, 27 Xinfa Road, Suzhou Industrial Park, Suzhou City, Jiangsu Province Patentee after: Feiyinuo Technology (Suzhou) Co.,Ltd. Address before: 215000 5th floor, building a, 4th floor, building C, 27 Xinfa Road, Suzhou Industrial Park, Suzhou City, Jiangsu Province Patentee before: VINNO TECHNOLOGY (SUZHOU) Co.,Ltd. |
|
| CP01 | Change in the name or title of a patent holder | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 215000 5th floor, building a, 4th floor, building C, 27 Xinfa Road, Suzhou Industrial Park, Suzhou City, Jiangsu Province Patentee after: Feiyinuo Technology Co.,Ltd. Address before: 215000 5th floor, building a, 4th floor, building C, 27 Xinfa Road, Suzhou Industrial Park, Suzhou City, Jiangsu Province Patentee before: Feiyinuo Technology (Suzhou) Co.,Ltd. |